INET Meeting Room, Manor Road Building
A market going over a 'tipping point' is the usual explanation of a market that has crashed. The formal interpretation is that markets lose stability as the current equilibrium point is annihilated and the market adjusts, via a non-equilibrium transition, to a new but quite different equilibrium. Although this is an old idea, dating back to at least the 1970's and Zeeman's qualitative approach, there is no clear evidence of the temporal formation and subsequent loss of stability of multi-modal distributions in housing market data. In this talk I will present the background of Zeeman's work and the extension to stochastic bi-modality of Cobb and Wagenmakers et al. Then, by extending Wagenmakers numerical techniques to the time evolution of the modes of housing market indices (1980 to present) I will show that there is clear evidence of multiple modes forming prior to the sub-prime crisis and that these modes collapsed just prior to the crisis. The results are compared to a number of other OECD countries where a large variety of dynamics are observed.